Factorial validity by tautology: Methodological comments on Stewart and Stewart
Dorans, N. J. (1977). Factorial Validity by Tautology: Methodological Comments on Stewart and Stewart. Multivariate Experimental Clinical Research, 3(2), 69-73.
A recent attempt to assess the factorial validity of the Eight State Questionnaire used an unusual methodological procedure that is tautological in nature. Retention and rotation of all principle components to a trivial simple structure, in which each component is marked by one and only one scale, is guaranteed for any nonsingular correlation matrix. The certain existence of the transformation to this trivial structure precludes the use of its existence as scientific evidence of factorial validity. A verbal, geometric argument and a formal mathematical proof support the contention that this unusual methodology is tautological, and hence should be avoided in future research.